Vision-Based Checkout Guidance With Tactile Tray and Audio Cues
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Vision-based self-checkout systems pose challenges for visually impaired individuals as they struggle to locate items on the checkout tray and perform necessary movements to facilitate item recognition, often requiring assistance from an attended checkout lane.
Innovation Solution
A vision-based self-checkout system equipped with multiple cameras, a tactile checkout tray with raised dots, and a universal navigation module with tactile buttons and audio interface, enabling visually impaired users to locate and identify items through audio guidance and tactile feedback.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a vision-based self-checkout system is used, then checkout automation and efficiency are improved, but accessibility for visually impaired users deteriorates
Solution Approach 1:
The patent introduces audio output devices and tactile feedback mechanisms as intermediaries between the vision-based detection system and visually impaired users. The system captures image data and generates audio descriptions of detected items, their positions, and checkout status, which are transmitted through audio output devices. Tactile buttons provide physical feedback for user interactions, serving as a mediator that translates visual information into accessible sensory modalities for visually impaired users.
Solution Approach 2:
The patent replaces the purely visual interaction mechanism with audio-based and tactile-based mechanisms. Instead of relying on visual displays and camera feedback that visually impaired users cannot access, the system substitutes these with audio output devices that provide verbal descriptions and tactile buttons that provide physical feedback, thereby substituting the mechanical visual interaction system with accessible sensory alternatives.
2Measurement precision
If multiple cameras are deployed for comprehensive item detection, then detection accuracy is improved, but system complexity increases
Solution Approach 1:
The patent divides the detection task among multiple cameras positioned at different locations and angles around the checkout area. Each camera captures a specific portion or perspective of the checkout scene, and the system processes images from each camera separately before integrating the results. This segmentation allows comprehensive item detection while maintaining manageable complexity by processing each camera's data independently.
Solution Approach 2:
The patent combines the detection results from multiple cameras to achieve comprehensive item recognition and tracking. By merging the data streams and detection outcomes from several camera sources, the system achieves complete coverage of the checkout area and accurate identification of all items, overcoming the limitations of any single camera while maintaining systematic processing.
3Measurement precision
If the vision system requires item movement for proper identification, then recognition accuracy is improved, but ease of use for visually impaired users deteriorates
Solution Approach 1:
The patent implements an audio feedback loop that provides real-time information to users about item detection status, recognition accuracy, and required actions. The system continuously monitors item positions, processes image data, and provides audio feedback describing which items are properly positioned for recognition and which require movement. This feedback mechanism guides visually impaired users in adjusting item positions without requiring them to visually assess the situation.
Solution Approach 2:
The patent introduces audio output devices as an intermediary that translates the vision system's detection status and positioning requirements into audible instructions for visually impaired users. The system acts as a mediator by processing visual information from cameras and converting it into audio descriptions that guide users on how to position items for optimal recognition, bridging the gap between the vision-based detection system and accessibility requirements.
Data Source
AI summary
A self-service checkout terminal for vision-based checkout includes a computing device having a processor and a memory. A camera coupled to the computing device has a predefined field of view focused on a scan zone and provides a video output signal of the scan zone to the computing device. A checkout tray has a tactile surface formed from a predefined number of rows and a predefined number of columns of raised dot. The checkout tray is positioned within the scan zone. The processor processes the video output signal from the camera to attempt to identify an item placed on the checkout tray by a user during a transaction, provides a description of the item to the user via the user interface when the item is identified, and provides the user of an indication of a location of the item via a user interface when the item is not identified.